aeo-audit
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill retrieves configuration data from localized paths within the user's home directory (~/.claude-marketing/). This access is tailored to its purpose of auditing brand context and does not involve sensitive system files or credentials.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. 1. Ingestion points: Context is loaded from JSON and Markdown files in ~/.claude-marketing/ (SKILL.md). 2. Boundary markers: Explicit delimiters are not used for external data. 3. Capability inventory: The skill uses the seo-specialist agent to process data and generate reports. 4. Sanitization: No specific sanitization or filtering logic is mentioned for the imported files.
Audit Metadata